Homelessness among veterans has remained a national problem for over three decades. Last year, Secretary of Veterans Affairs Eric Shinseki has pledged homelessness as one of his top 5 priorities and has announced a 5-year plan to end homelessness among veterans. One of the most important VA programs for homeless veterans, the Housing and Urban Development-Veterans Affairs Supportive Housing (HUD-VASH) program is in need of evaluation, research, and improvement. HUD-VASH is the largest supported housing program for veterans in the nation and is continuing to expand. The HUD-VASH program currently provides over 20,000 housing vouchers for rental assistance along with supportive case management services for adults with mental or physical disabilities who are chronically homeless; over 10,000 more housing vouchers have been approved this year. The objective of the proposed research is to investigate the real-world operation of the HUD-VASH program locally and nationally, with the goal of improving resource allocation and tailoring VA homeless services to veteran needs. There are three specific aims. The first aim is to evaluate an innovative group- intensive peer-support (GIPS) model of case management for HUD-VASH at the VA Connecticut Healthcare System. Performance and client outcomes at this study site will be compared to other control HUD-VASH sites during the same time using VA administrative program data. It is hypothesized that this model will be a cost- effective alternative model of case management for HUD-VASH and other supported housing programs across the country. The second aim is to conduct a national survey of HUD-VASH clinicians about their case management practices and attitudes about using groups in the program. The results of the survey will help establish a baseline for current case management practices and identify sites that may be interested in participating in a larger demonstration of the GIPS model. The third aim is to assess housing placement and program participation in HUD-VASH nationally. Factors related to the rapidity of housing placement and duration of program participation in HUD-VASH will be examined. Analysis of recent VA administrative program data will be used and the results will be informative in program development. This application is to support Jack Tsai, Ph.D., in conducting the proposed research and to help him develop a career as an independent investigator with a focus on VA homeless services. Dr. Tsai will work primarily under the mentorship of Robert Rosenheck, M.D., an internationally renowned mental health services researcher who has evaluated hundreds of homeless programs in the VA and established the first 40 HUD- VASH programs in the country. Dr. Tsai will be supported by a secondary mentor, James McGuire, Ph.D., who is a national administrative leader of VA homeless services, and a third mentor, Haiqun Lin, M.D., Ph.D., who is a biostatistician and an expert on longitudinal data analyses. An additional consultant, Vince Kane, MSW, has been recruited to provide administrative support and guidance on dissemination. Through this career development award, Dr. Tsai will be able to evaluate the main VA supported housing program for homeless veterans locally and nationally. Through coursework and mentorship, Dr. Tsai will gain training on research methodology and policy. He will develop expertise in program evaluation with the ultimate goal of improving homeless services based on research. At the end of the award, Dr. Tsai will be an established researcher in the VA and be in a position to compete for independent funding. PUBLIC HEALTH RELEVANCE: The proposed project will support Jack Tsai, Ph.D., in his career development and research on VA homeless services. The VA's flagship supported housing program for homeless veterans, the Housing and Urban Development-Veterans Affairs Supportive Housing (HUD-VASH) program, will be evaluated locally and nationally. A new group-intensive peer-support model of case management for the program will be tested and national interest in this new model will be surveyed. Factors related to program participation and housing placement will also be examined. Multivariate longitudinal data analyses will be employed and VA administrative program data will be used. This research is directly in line with VA priorities and the VA's goal t end homelessness among veterans. Findings from the project will guide future program development of VA homeless services. The results have the potential to improve program services for homeless veterans and identify areas in need of modification. The proposed project will also yield data for future funding proposals.